Hello! I've figured out what kind of database I have.
I read the manual, there is the following statement there: “In the future, all these databases will become available for HHblits in the HHblits format described in the previous subsection. For the time being, however, these databases are formatted for as they have been in the past: The eight databases marked by asterisks contain both HMMs in HHsearch format (.hhm.tar ) and the multiple sequence alignments (MSAs) in A3M format (.a3m.tar )”.
It turned out my database is in the old concatenated format. I did the following to check it: There are several *.hhm_dbs inside. I tried to rename these files and unpack them - they are not in *.tar format or *.tar.gz -format. These dabases cannot be unpacked with tar -vxf filename.tar or tar -xzvf filename.tar .gz. When I typed: file *hhm_db, I had :data. So they are not the archives at all.
The hhm-databases have to be somehow splitted into separated *.hhm-files. Would some person be so kind to explain to me how to deal with these hhm_dbs, pull out the hhm-files and finish with the sets of hhm-files if it is possible? I need your hint - how to turn the concatenate into the sets of separated hhm-files?
In the manual I found another statement: "The packed files <dbname>_a3m_db and <dbname>_hhm_db contain simply the concatenated A3M MSAs and HHMs, respectively, with a \0 character at the beginning of each file. They are therefore human-readable and are parsable for specific MSAs or models using tools such as grep or search functions in text editors (which however should be able to ignore the \0 character). The .index les contain indices to provide fast access to these two packed files. The a3m files are not needed for a single search iteration when no output MSA is requested." This statement is not very clear to me, sorry, it didn't help me to split the concatenate into separated hhm-files.
Many-many thanks for your help or advise! Sincerely, Nat
